Hello Corghi,
Corghi Stefano wrote:
26mb is a big database??
NO
I must consider to use my old copy of ms access!
You could use that Access file with Base - connecting to the mdb and
using the MS Jet engine for the data store actually. On Windows only.
Can someone help me for an app where data must be moved from notebook
to office environment.
You could simply create 2 Base odb files. One is your embedded data, the
other is a file that connects to your MySQL server.
You would then drag each table from the one to the other.
*Don't* do that - well, OK if you want to you can but for full 'saw buck
and a half' worth of records...no.
What is your data like would be my first question? Any binary stuff,
images...I'm assuming not.
So I would either:
use the SCRIPT command in the embedded hsqldb engine to generate HSQLdb
SQL create and insert statements for the whole database, all tables,
view definitions, etc. Then I would transform the text file with the SQL
commands for MySQL syntax difference and run it against the MySQL server.
-or -
If there is only a table or a few: dump each to a csv file.
Simple, say you have a table named Table1 you can do this.
open the SQL window ( Tools>SQL)
Execute the command
SELECT INTO TEXT 'table1.csv' * from "Table1"
When you execute that a file table1.csv will be created in the same
directory where the odb file is. It will use the default HSQLdb text
file settings.
For details on changing that to suit your needs please see the TEXT
TABLE section in the HSQLdb documentation, found at http://hsqldb.org
Then I would LOAD the csv file into server. (The documentation for any
server will cover this, search for LOAD DATA and you will find it)
<snip>
i have a problem with a database of 26 mb and a table of 142000
record.
It's quite impossible to work with it every database operation
require a
lot of time.
Can someone give me suggestion for this matter?
1 - GUI browsing performance is not good with embedded tables over a few
10's of thousands of records
2 - GUI filtering is
3 - GUI find record is not good
4 - GUI query is
5 - You should change the default settings for the Java runtime if you
are going to work with embedded databases. You want to start off with
about 128 meg for the vm.
6 - If you are using Windows. You *really* need to ensure that your
virtual memory manager is using s pre-allocated swap file. If you allow
the OS to grow the file, Base will allocate LOTS of little chunks and
the performance is terrible, if it is also allocating the space on disk.
So using a pre-allocated swap file makes a big difference.
7 - COMPACT often. Tools>SQL in the window enter the command SHUTDOWN
COMPACT. When its done close the odb file. Do this on a schedule with a
frequency commensurate for the amount inserts or edits you are doing to
the data. If you are editing hundreds of records a day, then do this daily.
8 - If you are running Windows and using Base as a primary tool. Set the
system preferences to favor background processes over the GUI, this helps.
9 - Did I mention use SQL over browsing. Which means really use FILTERS
in the GUI if you do you will be surprised.
10 - start up is start up, it's slow..yeah - but doing that will make a
difference.
Best wishes,
Drew
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@dba.openoffice.org
For additional commands, e-mail: users-h...@dba.openoffice.org
